CoChat
Sign Up
PreprintCited by 1Academic Publishing and Open AccessConferences and Exhibitions ManagementPublishing and Scholarly Communication

Position: The Current AI Conference Model is Unsustainable! Diagnosing the Crisis of Centralized AI Conference

Open access
Chen, Nuo, Duan, Moming, Lin, Andre Huikai, Wang, Qian, Wu, Jiaying, He, Bingsheng·2025·arXiv
DOI 10.48550/arxiv.2508.04586arXiv 2508.04586
Open this paper in CoChat — free accountOpen PDF

Abstract

Artificial Intelligence (AI) conferences are essential for advancing research, sharing knowledge, and fostering academic community. However, their rapid expansion has rendered the centralized conference model increasingly unsustainable. This paper offers a data-driven diagnosis of a structural crisis that threatens the foundational goals of scientific dissemination, equity, and community well-being. We identify four key areas of strain: (1) scientifically, with per-author publication rates more than doubling over the past decade to over 4.5 papers annually; (2) environmentally, with the carbon footprint of a single conference exceeding the daily emissions of its host city; (3) psychologically, with 71% of online community discourse reflecting negative sentiment and 35% referencing mental health concerns; and (4) logistically, with attendance at top conferences such as NeurIPS 2024 beginning to outpace venue capacity. These pressures point to a system that is misaligned with its core mission. In response, we propose the Community-Federated Conference (CFC) model, which separates peer review, presentation, and networking into globally coordinated but locally organized components, offering a more sustainable, inclusive, and resilient path forward for AI research.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

DataCite · 10.48550/arxiv.2508.04586

Cites

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Submissions have grown sharply in recent years while the pool of reviewers with the necessary time and subfield expertise has not, and researchers increasingly report declining review quality.

    “That standard is becoming harder to meet, as submissions have grown sharply in recent years while the pool of reviewers with the necessary time and subfield expertise has not, and researchers increasingly report declining review quality [Chen et al., 2025, Powell, 2016].”

    From On the limits and opportunities of AI reviewers: Reviewing the reviews of Nature-family papers with 45 expert scientists (2026) · cites this paper as Chen et al., 2025

    The Introduction (p. 1) says rapid submission growth has overwhelmed peer review and often compromised review quality; the discussion on p. 10 also says resubmissions strain reviewer capacity. The Reddit discussions (p. 21) include a complaint that reviewers are not spending enough time, but the copy does not establish that the pool of reviewers with sufficient time and subfield expertise has failed to grow, or that reports of declining quality are increasing. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

  • Manuscript submission volumes continue to rise across both journals and conferences while the pool of qualified human reviewers does not expand at the same rate, leading to reviewer fatigue, longer review timelines, and difficulty recruiting reviewers willing to provide thorough evaluations.

    “Manuscript submission volumes continue to rise across both journals and conferences while the pool of qualified human reviewers does not expand at the same rate, leading to reviewer fatigue, longer review timelines, and difficulty recruiting reviewers willing to provide thorough evaluations”

    From On the limits and opportunities of AI reviewers: Reviewing the reviews of Nature-family papers with 45 expert scientists (2026) · cites this paper as Chen et al., 2025

    Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

No references could be resolved yet.

Research works printed but not linked: 30

  1. [1]1. Jing Yang. Position: The artificial intelligence and machine learning community should adopt a more transparent and regulated peer review process, 2025. URL https://arxiv.org/abs/2502. 00874.
  2. [2]2. Jaeho Kim, Yunseok Lee, and Seulki Lee. Position: The AI conference peer review crisis demands author feedback and reviewer rewards. In Forty-second International Conference on Machine Learning Position Paper Track , 2025. URL https://openreview.net/forum?id= l8QemUZaIA.
  3. [3]3. Sebastian Jäckle. The carbon footprint of travelling to international academic conferences and options to minimise it. In Kristian Bjørkdahl and Adrian Santiago Franco Duharte, editors, Academic Flying and the Means of Communication , pages 19-52. Springer Nature Singapore, Singapore, 2022. ISBN 978-981-16-4911-0. doi: 10.1007/978-981-16-4911-0\_2. · DOI
  4. [4]4. Andrea Gokus, Knud Jahnke, Paul M Woods, Vanessa A Moss, Volker Ossenkopf-Okada, Elena Sacchi, Adam R H Stevens, Leonard Burtscher, Cenk Kayhan, Hannah Dalgleish, Victoria Grinberg, Travis A Rector, Jan Rybizki, and Jacob White. Astronomy's climate emissions: Global travel to scientific meetings in 2019. PNAS Nexus , 3(5):pgae143, 04 2024. ISSN 2752-6542. doi: 10.1093/pnasnexus/pgae143. URL https://doi.org/10.1093/pnasnexus/ pgae143. · DOI
  5. [5]5. Helen Nicholls, Matthew Nicholls, Sahra Tekin, Danielle Lamb, and Jo Billings. The impact of working in academia on researchers' mental health and well-being: A systematic review and qualitative meta-synthesis. PLOS ONE , 17(5):1-29, 05 2022. doi: 10.1371/journal.pone. 0268890. URL https://doi.org/10.1371/journal.pone.0268890. · DOI
  6. [6]6. Tammy Mackenzie, Sreyoshi Bhaduri, Leslie Salgado, peer herholz, zachary rosenthal, Rubaina Khan, Animesh Paul, and Debarati Basu. Reimagining ai conference mission statements to promote inclusion in the emerging institutional field of ai, 2024.
  7. [7]7. Daniel Zhang, Christie Lawrence, Michael Sellitto, Russell Wald, Marietje Schaake, Daniel E Ho, Russ Altman, and Andrew Grotto. Enhancing international cooperation in ai research: The case for a multilateral ai research institute. Stanford Institute for Human-Centered Artificial Intelligence , pages 2022-05, 2022.
  8. [8]8. NeurIPS. Neural Information Processing Systems. https://neurips.cc/, 1986.
  9. [9]9. ICLR. International Conference on Learning Representations. https://iclr.cc/, 2013.
  10. [10]10. ICML. International Conference on Machine Learning. https://icml.cc/, 1980.
  11. [11]11. Caiming Zhang and Yang Lu. Study on artificial intelligence: The state of the art and future prospects. Journal of Industrial Information Integration , 23:100224, 2021. ISSN 2452-414X. doi: https://doi.org/10.1016/j.jii.2021.100224. URL https://www.sciencedirect.com/science/ article/pii/S2452414X21000248. · DOI
  12. [12]12. Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D Manning, Stefano Ermon, and Chelsea Finn. Direct preference optimization: Your language model is secretly a reward model. In Thirty-seventh Conference on Neural Information Processing Systems , 2023. URL https: //openreview.net/forum?id=HPuSIXJaa9.
  13. [13]13. Diederik P. Kingma and Jimmy Ba. Adam: A method for stochastic optimization. In The Third International Conference on Learning Representations , 2015. URL https://arxiv.org/abs/1412. 6980.
  14. [14]14. NeurIPS Conference. Guidance for NeurIPS Workshop Proposals 2025. https://neurips.cc/ Conferences/2025/CallForWorkshopsGuidance, 2025.
  15. [15]15. Dinghuai Zhang and Yuanqi Du and Guan-Horng Liu and Chenlin Meng and Ruiqi Gao and Max Welling and Yoshua Bengio . ICML Workshop: Structured Probabilistic Inference and Generative Modeling. https://icml.cc/virtual/2024/workshop/29946, 1980.
  16. [17]17. NeurIPS Conference. NeurIPS 2024 Fact Sheet. https://media.neurips.cc/Conferences/ NeurIPS2024/NeurIPS2024-Fact\_Sheet.pdf, 2024.
  17. [21]21. Albert Gu and Tri Dao. Mamba: Linear-time sequence modeling with selective state spaces, 2024. URL https://openreview.net/forum?id=AL1fq05o7H.
  18. [22]22. David Tran, Alex Valtchanov, Keshav Ganapathy, Raymond Feng, Eric Slud, Micah Goldblum, and Tom Goldstein. An open review of openreview: A critical analysis of the machine learning conference review process, 2020. URL https://arxiv.org/abs/2010.05137. · arXiv
  19. [23]23. Eduardo Araujo Oliveira, Madhavi Mohoni, Sonsoles López-Pernas, and Mohammed Saqr. Human-ai collaboration or academic misconduct? measuring ai use in student writing through stylometric evidence, 2025. URL https://arxiv.org/abs/2505.08828. · arXiv
  20. [25]25. Chiamaka Ann Marie Ajufo and Girish Bekaroo. An automated personal carbon footprint calculator for estimating carbon emissions from transportation use. In Proceedings of the International Conference on Artificial Intelligence and Its Applications , icARTi '21, New York, NY, USA, 2021. Association for Computing Machinery. ISBN 9781450385756. doi: 10.1145/ 3487923.3487935. URL https://doi-org.libproxy1.nus.edu.sg/10.1145/3487923.3487935. · DOI
  21. [27]27. Åsa Knaggård, Barry Ness, and David Harnesk. Finding an academic space: reflexivity among sustainability researchers. Ecology and Society , 23(4), 2018. ISSN 17083087. URL https: //www.jstor.org/stable/26796866.
  22. [28]28. Thomas Kwa, Ben West, Joel Becker, Amy Deng, Katharyn Garcia, Max Hasin, Sami Jawhar, Megan Kinniment, Nate Rush, Sydney Von Arx, et al. Measuring ai ability to complete long tasks. arXiv preprint arXiv:2503.14499 , 2025. · arXiv
  23. [29]29. Toby Ord. Is there a half-life for the success rates of ai agents? arXiv preprint arXiv:2505.05115 , 2025. · arXiv
  24. [30]30. NeurIPS Conference. Announcement: We're excited to announce that neurips2025 will also be held in mexico city! https://x.com/NeurIPSConf/status/1866605636491801021, 2025.
  25. [31]31. NeurIPS Conference. Announcement: We're excited to announce that neurips2025 will be held in copenhagen! https://x.com/NeurIPSConf/status/1866605640145064216, 2025.
  26. [33]33. NeurIPS Conference. NeurIPS 2024 Registration Changes. https://blog.neurips.cc/2024/10/ 29/neurips-2024-registration-changes/, 2024.
  27. [34]34. Reddit. NeurIPS is pushing to SACs to reject already accepted papers due to venue constraints. https://www.reddit.com/r/MachineLearning/comments/1n4bebi/d\_neurips\_is\_ pushing\_to\_sacs\_to\_reject\_already/, 2025.
  28. [35]35. KDD. KDD 2026 Submission Guidelines. https://kdd2026.kdd.org/research-track-call-forpapers/, 2025.
  29. [41]41. Peter Scarborough, Paul N Appleby, Anja Mizdrak, Adam D M Briggs, Ruth C Travis, Kathryn E Bradbury, and Timothy J Key. Dietary greenhouse gas emissions of meat-eaters, fish-eaters, vegetarians and vegans in the uk, 2014. URL https://pmc.ncbi.nlm.nih.gov/articles/ PMC4372775/.
  30. [43]43. Viktoria Mannheim and Judit Lovasné Avató. Examining the carbon footprint of conferences with an emphasis on energy consumption and catering, 2025. URL https://www.mdpi.com/ 1996-1073/18/2/321.

Cited by

Sorted by citations

Authors
Chen, N., Duan, M., Lin, A. H., Wang, Q., Wu, J., & He, B.
ORCID
Venue
arXiv
Publisher
arXiv
Published
2025
ISSN
2331-8422 (ISSN-L)
DOI
10.48550/arxiv.2508.04586
arXiv
2508.04586
OpenAlex
W6929596009
Cited by
0
References
0
Type
posted content
Language
en
Open access
Licence
creativecommons.org/licenses/by/4.0/legalcode
Locations
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University) (cc-by)
Topics
  • Social Sciences › Decision Sciences › Information Systems and Management › Academic Publishing and Open Access
  • Social Sciences › Sociology and Political Science › Conferences and Exhibitions Management
  • Social Sciences › Arts and Humanities › History and Philosophy of Science › Publishing and Scholarly Communication
Keywords
conference sustainability, peer review, scientific dissemination

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review